Quant · Core
Information Ratio
The information ratio is active return over active risk — residual performance per unit of tracking error versus a benchmark.
Definition
IR = (r_p − r_b) / TE. It is the Sharpe of the residual. Grinold’s fundamental law: IR ≈ IC × √breadth if you believe the fairy tale and ignore costs.
Why it matters
Active mandates are hired and fired on IR. A high IR on a closet indexer is a rounding error times a small TE.
Case
A manager with 50bp TE and 25bp excess: IR = 0.5. Another with 400bp TE and 80bp excess: IR = 0.2. The first looks ‘skilled’ and did not take enough risk to matter.
How to read it
TE definition, after-cost excess, and whether the benchmark is the actual mandate. IR on a 14-month sample is noise.
No query has been run yet. Which is tragically normal for most knowledge systems, but we are trying to evolve past decorative databases.